Maximizing Lifetime in Heterogeneous Sensor Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Maximizing Lifetime in Heterogeneous Sensor Networks

Résumé

Due to technology advancements in fields such as miniaturization, that enabled their deployment in several and diverse contexts, Wireless Sensor Networks (WSNs) have been object of intense research in the last few years. They find use for example in military, environmental control and healthcare applications. WSNs are composed of several small devices, usually equipped with non-rechargeable batteries, which can be used together to perform complex monitoring activities. However, since advances in energy storage have been less impressive, optimizing the battery consumption is essential to carry on such monitoring activities for as long as possible. In this work we consider a heterogeneous scenario where a series of points of interests scattered over a given geographical region are monitored by sensors belonging to two different families; namely, omnidirectional sensors that monitor targets located within a certain distance threshold from them in any direction, and directional ones (such as photo or video cameras) that only monitor a limited angle, and can freely rotate around themselves. We also consider the possibility of special nodes mounting two sensors (one for each type) powered by a single battery. In the above described scenario, our aim is to individuate several non-disjoint subsets of sensors (called covers) that can independently monitor all targets with both families, as well as appropriate activation times for each cover, such that the monitoring activity can be performed for as long as possible and the battery lifetime of each individual sensor is not violated. To this end, we propose an exact algorithm based on Column Generation. Since the subproblem is a specialization of the Set Covering problem and is therefore NP-Hard, we combined it with a memetic metaheuristic algorithm (MA). Our MA is able to provide several good solutions (covers) in a single iteration, bringing a noticeable speed-up in particular at the beginning of the search phase. The subproblem is solved to optimality in the final iterations, when the MA fails to provide attractive covers and to eventually prove optimality of the current solution. The effectiveness of the proposed approach is proven on a wide set of test instances.
Fichier non déposé

Dates et versions

hal-01067841 , version 1 (24-09-2014)

Identifiants

  • HAL Id : hal-01067841 , version 1

Citer

Andrea Raiconi, Raffaele Cerulli, Monica Gentili, André Rossi, Marc Sevaux. Maximizing Lifetime in Heterogeneous Sensor Networks. Proceedings of the 44th Annual Conference of the Italian Operational Research Society (AIRO 2014), Sep 2014, Como, Italy. ⟨hal-01067841⟩
112 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More